skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Characterizing Complexity of Containerized Cargo X-ray Images

Abstract

X-ray imaging can be used to inspect cargos imported into the United States. In order to better understand the performance of X-ray inspection systems, the X-ray characteristics (density, complexity) of cargo need to be quantified. In this project, an image complexity measure called integrated power spectral density (IPSD) was studied using both DNDO engineered cargos and stream-of-commerce (SOC) cargos. A joint distribution of cargo density and complexity was obtained. A support vector machine was used to classify the SOC cargos into four categories to estimate the relative fractions.

Authors:
 [1];  [1];  [1];  [1];  [1]
  1. Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
Publication Date:
Research Org.:
Lawrence Livermore National Lab. (LLNL), Livermore, CA (United States)
Sponsoring Org.:
USDOE; US Dept. of Homeland Security (DHS). Domestic Nuclear Detection Office
OSTI Identifier:
1305809
Report Number(s):
LLNL-TR-700977
TRN: US1601803
DOE Contract Number:
AC52-07NA27344; AC05-06OR23100; IAA HSHQDC-12-X-00341
Resource Type:
Technical Report
Country of Publication:
United States
Language:
English
Subject:
46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND TECHNOLOGY; X-RAY SPECTRA; CARGO; DENSITY; IMPORTS; SPECTRAL DENSITY; IMAGES; X RADIATION; DISTRIBUTION; INSPECTION; PERFORMANCE

Citation Formats

Wang, Guangxing, Martz, Harry, Glenn, Steven, Divin, Charles, and Birrer, Nat. Characterizing Complexity of Containerized Cargo X-ray Images. United States: N. p., 2016. Web. doi:10.2172/1305809.
Wang, Guangxing, Martz, Harry, Glenn, Steven, Divin, Charles, & Birrer, Nat. Characterizing Complexity of Containerized Cargo X-ray Images. United States. doi:10.2172/1305809.
Wang, Guangxing, Martz, Harry, Glenn, Steven, Divin, Charles, and Birrer, Nat. Fri . "Characterizing Complexity of Containerized Cargo X-ray Images". United States. doi:10.2172/1305809. https://www.osti.gov/servlets/purl/1305809.
@article{osti_1305809,
title = {Characterizing Complexity of Containerized Cargo X-ray Images},
author = {Wang, Guangxing and Martz, Harry and Glenn, Steven and Divin, Charles and Birrer, Nat},
abstractNote = {X-ray imaging can be used to inspect cargos imported into the United States. In order to better understand the performance of X-ray inspection systems, the X-ray characteristics (density, complexity) of cargo need to be quantified. In this project, an image complexity measure called integrated power spectral density (IPSD) was studied using both DNDO engineered cargos and stream-of-commerce (SOC) cargos. A joint distribution of cargo density and complexity was obtained. A support vector machine was used to classify the SOC cargos into four categories to estimate the relative fractions.},
doi = {10.2172/1305809},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Fri Aug 19 00:00:00 EDT 2016},
month = {Fri Aug 19 00:00:00 EDT 2016}
}

Technical Report:

Save / Share:
  • X-ray inspection systems can be used to detect radiological and nuclear threats in imported cargo. In order to better understand performance of these systems, the attenuation characteristics of imported cargo need to be determined. This project focused on developing image processing algorithms for segmenting cargo and using x-ray attenuation to quantify equivalent steel thickness to determine cargo density. These algorithms were applied to over 450 cargo radiographs. The results are summarized in this report.
  • The objective of this report is to present information in both graphical form and data sheet form for radiographers to use as a guide for characterizing any x-ray machine for any given material capable of being x-rayed. This was accomplished by characterizing the Philips MG 301 Constant Potential X-ray Machine for radiographing various steel thicknesses with a 4.0mm focal spot. The information in this report will permit users of any x-ray machine to develop exposure charts to allow them to make fast and reliable decisions when faced with the problem of developing an exposure technique for radiographing, for example, thicknessesmore » of steel in the range of .125 to 2.5 inches. Information contained in this report provides users of the Philips MG 301 and other types of x-ray machines with guidelines for developing exposure techniques for other metals (aluminum, copper, etc.) in the thickness range stated above. This can be accomplished by using steel as a cross reference in conjunction with recorded charts of approximate radiographic equivalence factors.« less
  • The ability to quickly and accurately characterize arbitrary multilayers is very valuable for not only can we use the characterizations to predict the reflectivity of a multilayer for any soft x-ray wavelength, we also can generalize the results to apply to other multilayers of the same type. In addition, we can use the characterizations as a means of evaluating various sputtering environments and refining sputtering techniques to obtain better multilayers. In this report we have obtained improved characterizations for sample molybdenum-silicon and vanadium-silicon multilayers. However, we only examined five crystals overall, so the conclusions that we could draw about themore » structure of general multilayers is limited. Research involving many multilayers manufactured under the same sputtering conditions is clearly in order. In order to best understand multilayer structures it may be necessary to further refine our model, e.g., adopting a Gaussian form for the interface regions. With such improvements we can expect even better agreement with experimental values and continued concurrence with other characterization techniques. 18 refs., 30 figs., 7 tabs.« less
  • We present a general statistical technique for characterizing x-ray sensitive linear diode arrays and CCD arrays. We apply this technique to characterize the response of a linear diode array, Princeton Instrument model X-PDA, and a virtual phase CCD array, TI 4849, to direct illumination by x-rays. We find that the response of the linear array is linearly proportional to the incident intensity and uniform over its length to within 2 %. Its quantum efficiency is 38 % for Cu K{sub {alpha}} x-rays. The resolution function is evaluated from the spatial autocorrelation function and falls to 10 % of its peakmore » value after one pixel. On the other hand, the response of the CCD detecting system to direct x-ray exposure is non-linear. To properly quantify the scattered x-rays, one must correct for the non- linearity. The resolution is two pixels along the serial transfer direction. We characterize the noise of the CCD and propose a model that takes into account the non-linearity and the resolution function to estimate the quantum efficiency of the detector. The quantum efficiency is 20 %.« less
  • For light curves sampled on an uneven grid of observation times, the shape of the power density spectrum (PDS) includes severe distortion effects due to the window function, and simulations of light curves are indispensable to recover the true PDS. We present an improved method for comparing light curves generated from a PDS model to the measured data and apply it to a 50-day long RXTE observations of NGC 4945, a Seyfert 2 galaxy with well-determined mass from megamaser observations. The improvements over previously reported investigations include the adjustment of the PDS model normalization for each simulated light curve inmore » order to directly investigate how well the chosen PDS shape describes the source data. We furthermore implement a robust goodness-of-fit measure that does not depend on the form of the variable used to describe the power in the periodogram. We conclude that a knee-type function (smoothly broken power law) describes the data better than a simple power law; the best-fit break frequency is {approx} 10{sup -6} Hz.« less